Application of Wavelet Energy Feature in Facial Expression Recognition

This paper proposes a high performance facial expression recognition method based on wavelet energy feature (WEF). As wavelet energy feature can discriminate the texture of expression images, it is used in facial expression recognition for the first time. Fisher's linear discriminants (FLD) can describe the details of the image, so we combine FLD with WEF. WEF is added to the image first, and then FLD is used to feature extraction. Finally, we use the Nearest-Neighbor rule to classify the seven expressions (anger, disgust, fear, happiness, normal, sadness, surprise) of JAFFE. The very high recognition rate obtained in experiments shows the effect of the proposed method.

[1]  Z. Zenn Bien,et al.  Facial emotional expression recognition with soft computing techniques , 2005, ROMAN 2005. IEEE International Workshop on Robot and Human Interactive Communication, 2005..

[2]  Ioannis Pitas,et al.  ICA and Gabor representation for facial expression recognition , 2003, Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429).

[3]  M. Grgic,et al.  Statistics in face recognition: analyzing probability distributions of PCA, ICA and LDA performance results , 2005, ISPA 2005. Proceedings of the 4th International Symposium on Image and Signal Processing and Analysis, 2005..

[4]  Liu Chong-qing.  Research on face recognition method based on holistic facial features , 2003 .

[5]  Maja Pantic,et al.  Automatic Analysis of Facial Expressions: The State of the Art , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Dae-Jong Lee,et al.  Emotion recognition from the facial image and speech signal , 2003, SICE 2003 Annual Conference (IEEE Cat. No.03TH8734).